Instructor: Connie Figueroa Schibber
This course covers statistical modeling with explicitly defined hierarchies. Social scientists encounter multilevel data all the time: voters clustered in electoral districts, students nested within classrooms, legislators clustered in congressional periods, countries nested within regions, and so forth. Classic time-series cross-sectional (TSCS) data can also be thought as multilevel data, with observations clustered by unit and time period. In survey research, multilevel regression and poststratification (MRP) is a method to estimate public opinion across geographic units from individual-level survey data.
The course will focus on multilevel nested models and multilevel non-nested models for linear and generalized linear models. It will feature frequentist and Bayesian perspectives on inference and computation of hierarchical models.
A working syllabus is available here.